incorrect result from dpnp.matmul when out keyword is f_contiguous

[vtavana]

dpnp.matmul returns the data in the result array in an incorrect order when out keyword is given and it is f_contiguous.
For instance,

>>> import dpnp
>>> a=dpnp.arange(4.).reshape(2,2)
>>> b=dpnp.array(a, order="C")
>>> out = dpnp.empty((2,2), order="F")
>>> dpnp.matmul(b,b, out=out)
# array([[ 2.,  6.],
#       [ 3., 11.]], dtype=float32)

>>>   out.flags
# C_CONTIGUOUS : False
# F_CONTIGUOUS : True

>>> import numpy
>>> a=numpy.arange(4.).reshape(2,2)
>>> b=numpy.array(a, order="C")
>>> out = numpy.empty((2,2), order="F")
>>> numpy.matmul(b,b, out=out)
# array([[ 2.,  3.],
#        [ 6., 11.]])

>>>   out.flags
# C_CONTIGUOUS : False
# F_CONTIGUOUS : True

Or for higher dimensions:

>>> import dpnp
>>> a=dpnp.arange(8.).reshape(2,2,2)
>>> b=a.transpose(0, 2, 1)
>>> out = dpnp.empty((2,2,2))
>>> out=out.transpose(0, 2, 1)
>>> dpnp.matmul(b,b, out=out)
# array([[[ 2.,  3.],
#        [ 6., 11.]],
#       [[46., 55.],
#        [66., 79.]]], dtype=float32)

>>> import numpy
>>> a=numpy.arange(8.).reshape(2,2,2)
>>> b=a.transpose(0, 2, 1)
>>> out = numpy.empty((2,2,2))
>>> out=out.transpose(0, 2, 1)
>>> numpy.matmul(b,b, out=out)
# array([[[ 2.,  6.],
#        [ 3., 11.]],
#       [[46., 66.],
#        [55., 79.]]])